Genome-wide association studies (GWAS) have identified common polymorphisms at the CXCL12 locus significantly associated with coronary heart disease (CHD). We have reproduced the association of this locus with CHD risk in multiethnic studies and have identified a novel variant that increases CHD risk, decreases plasma CXCL12 levels and alters expression of CXCL12 gene. We further hypothesized that if CXCL12 is causally involved in the pathogenesis of CHD than genetic variation at its primary receptor CXCR4 may also be associated with CHD. To test our hypothesis, we investigated genetic variation at the CXCR4 locus through meta-analyses of genetic data on common variants involving ~63,000 CHD cases and ~92,000 controls and fine-mapping experiments in ~50,000 CHD cases and ~50,000 controls. First, we were able to identify a common variant at the CXCR4 locus that is strongly associated with CHD risk (P = 4x10-7). Second, preliminary analyses of fine-mapping studies identified: (i) a low frequency variant in the promoter region of CXCR4 that associates with CHD risk at experiment-wide significance levels (OR: 1.13; P-value: 7.1x10-6) and (ii) a naturally occurring rare missense variant in the CXCR4 gene (I57L; OR: 2.16 P = 6x10-3) associated with increased CHD risk. The same analyses also identified a low-frequency missense variant in the CXCL12 gene (R125C; OR: 3.38; P = 0.01) nominally associated with increased CHD risk. Through mechanistic studies, we were able to further demonstrate that endothelial specific deficiency of CXCR4 leads to accelerated atherosclerosis in mice. Capitalizing on these findings, we plan to conduct genetic, biomarker, and mechanistic studies to investigate the underlying biological mechanisms and the directional impact of the CXCL12/CXCR4 pathway on disease risk. Such evidence will help prioritize or deprioritize existing therapeutic programs that are already considering CXCL12 as a potential therapeutic target in CHD. In particular, we will reinforce our ongoing fine-mapping experiments by integrating (i) whole-exome sequencing studies in 10,000 early-onset MI cases (men and women aged ? 50 years) and 10,000 controls. We will further conduct (ii) measurements for circulating plasma CXCL12 levels in 12,000 participants (including 5,500 with incident CHD events and 3,000 participants in a high risk CHD group) enrolled in three prospective studies and three clinical trials. We will integrate evidence on plasma CXCL12 levels, genes and CHD risk to help assess directionality and utility of plasma CXCL12 as a marker of causality. We will mechanistically evaluate at the CXCL12 locus the lead coding and non- coding variants associated with CHD risk in human and murine cells and on mouse atherosclerosis and myocardial infarction. We will also mechanistically evaluate CXCR4 genetic variation in human iPS cell derived macrophages and endothelial cells. We will obtain genotype specific human iPS cells through two complimentary approaches: (i) recall of participants based on genotype and (ii) targeted genome editing through TALENs. We will further conduct studies to investigate CXCR4 expression and protein function by genotype through specific quantitative and functional assays. Findings from the proposed study should have considerable translational implications. The two joint-PIs provide internationally recognized cross-disciplinary expertise in vascular biomarkers, epidemiology, experimental biology, and clinical medicine, a combination of skills that will help translate findings for clinical benefit.